Profitable Trade-Off Between Memory and Performance In Multi-Domain Chatbot Architectures

TL;DR

This paper explores a trade-off between memory usage and performance in multi-domain chatbot architectures by comparing a single, multi-task BERT model with multiple specialized models, aiming to optimize server load and classification accuracy.

Contribution

It introduces a masking method for multi-task BERT classification, evaluating its effectiveness against specialized models across diverse datasets in a chatbot context.

Findings

Single BERT model with masking performs comparably to multiple models.

Multi-task BERT reduces server load and memory usage.

Performance varies depending on dataset complexity.

Abstract

Text classification problem is a very broad field of study in the field of natural language processing. In short, the text classification problem is to determine which of the previously determined classes the given text belongs to. Successful studies have been carried out in this field in the past studies. In the study, Bidirectional Encoder Representations for Transformers (BERT), which is a frequently preferred method for solving the classification problem in the field of natural language processing, is used. By solving classification problems through a single model to be used in a chatbot architecture, it is aimed to alleviate the load on the server that will be created by more than one model used for solving more than one classification problem. At this point, with the masking method applied during the estimation of a single BERT model, which was created for classification in more than one subject, the estimation of the model was provided on a problem-based basis. Three separate data sets covering different fields from each other are divided by various methods in order to complicate the problem, and classification problems that are very close to each other in terms of field are also included in this way. The dataset used in this way consists of five classification problems with 154 classes. A BERT model containing all classification problems and other BERT models trained specifically for the problems were compared with each other in terms of performance and the space they occupied on the server.